I am a final year student at Hampshire College and instead of the ususal classes each student does a year long project. So I will be working on a singe "game" all year long and I have several ideas as to what I would like to to. My question is this, on what level should I develop my game. I have been told to write my own engine, since coding skills are prized above game design. Every one has "this great idea for a game". I have also been told that since game engines are converging rapidly and most features are fairly well explored and/or available in a library now a days that a new game engine would be re-inventing the wheel. They suggested taking an off the shelf engine, like Torque, Orge, or Cube and really fleshing out an actual game. I think this has a bit of merit to it. I could design the hell out of a game down to exacting detail use those documents as my final project. Basically do my project in game design instead of implementation. I could also design some interesting "sysytem" that would be appropriate to a game. I have toyed with the idea of creating a system for backround units to swordfight. Units in turn based fantasy games could actually dual with each other instead of stand around and swing their weapon every 10 seconds. --------------------- So what are peoples thoughts, what skills are more interesting and sought after by employers? Thanks in advance, Mikel

Like the previous poster stated it all depends what you want. I would suggest on doing a project that best displays your strenghts as a programmer, be it AI, physics ect. Just try to set out a nice calendar with milestone dates ect, so you can keep yourself in check and make sure you meet those milestones.

Taking an engine like torque has the advantages of giving you more time to do actual game mechanics stuff but the disadvantage that 90% of the core engine is done for you.

You have a year, sit back and think what you can do with the time given to you, but dont try to make unreachable goals either.

It really depends on a couple of factors. You never stated what major you are, are you Computer Science or something else? When you plan on working in the industry what do you want to do?

Have you written any games before? Do you have any experience creating mods or conversions? Are you working on a team or by yourself?
What are your strengths? Do you do great software engineering/design or can you implement just about anything? do you have any experience using OpenGL/Direct3D? How about sound?


I'm a CS Major who wants to program in the game industry. Because of these factors I decided to develop my own engine for my senior project, I felt it would give me much more experience in the problem domain than using something that already exists. I learned a great deal through the design and development and am very glad I made that choice.

I am working on another game right now for an independant study for which I am also developing an engine from scratch. I'm not doing this because it will look better or have cooler features, but because there are certain things I want to learn and this is a fun and challenging way to do that.

Sometime in the future I would like to play around with other peoples engines to learn certain aspects of how they solved problems and how other people have setup their systems. For me more knowledge sticks if I find a solution and then learn other peoples solutions rather than just having the answer available. That way I already have knowledge of the problem domain, and I'm just learning other peoples solutions. If you learn the really good solutions to the problems before you find the problems I think you loose a certain sense of depth in the domain.

When you make the choice between using an off-the-shelf engine or creating your own I would really recommend selecting the option that will teach you the skills you want to learn. Otherwise what's the point?

First off, Thanks.

My school has no "majors", for this section of my education I write a contract and that defines what I am responsible for.

I have experience with OpenGl mostly and have solid c++ skills. I have some Software engineering experience and have taken a couple AI classes as well. At this point I have been mostly unfocused.

I have tried a couple times to write a game engine but I had not had the math, data strucuture or s.e. experience and ran into problems that I could not solve. Mostly they were design related.

I intend to set very specific goals and deadlines for myself all of which will be in the contract.

I think the best thing for me is to write my own engine, using as much of other peoples code as possible. that way I can get a feel for the issues involved with out having to write a sound or physics library from scratch.

I worry that I will produce something technically sound with some great features but that will look kind of blah. A well designed game without nice textures or well placed sound does not look astounding compared to a torque game. But I guess learning is really my goal so that is what I should go for.

Thanks y'all

And whatever you do make sure it works well immediately. SOP is to scan the disk for viri and give it one shot to wow. If it doesn't work "out of the box" you probably won't get a second chance unless the words on your resume are impressive (actually you most likely won't get the interview with an impressive demo unless the words look good too - use a spell checker).

It is very competitive. During one of my previous interviews they told me that I was in the top 50 for the position. Think about it.

Quote:You also might want to consider how many companies need someone who can write an engine from scratch as opposed to how many companies need someone who can work with pre-existing engines.

This is a good point. And consider this; There are alot more applications that go into making a game than the game engine. The tools can be quiet complex as well. Besides 3d/2d art packages additional editors are needed for levels, npc's, objects, etc...

These can also be a bit more impressive to an academic type (ie your professor) since a well implemented gui application LOOKS impressive (menubar, toolbars, dialogs, etc...) but in fact gui toolkits (qt, gtk, wxwidgets, mfc, java swing etc) can greatly simplify/ease developement.

Another thing to consider: game demos for programmers are not really required, at least not in my experience. A lot of companies that I interviewed with, including the one that hired me, never even asked to see my demo game. (even when I put so much work into it!).

And now that I help out with hiring, I get to look at people's resumes, and we check out the demos every once in a while, but never put much faith in them. Because a) you don't know how much of the games code was actually written by that person, and b) you don't know how long it took (if some guy hands us an awesome version of tetris, but it took him a year to write it, then he's not a good hire)

Writing a demo game is still very valuable, if for nothing else than educational value. And there are still some companies that want to see it. And there's all those jobs that require "previous experience" (and writing a demo game sorta counts as previous experience).

Anyway, my point is: you might want to think more about projects that will be educational for you to write, instead of projects that will look good to others.